Jabiru engine failures
Join Date: May 2013
Location: have I forgotten or am I lost?
Age: 71
Posts: 1,126
Likes: 0
Received 0 Likes
on
0 Posts
jabiru aren't unique in experiencing engine problems.
crankcase through bolts used to fail on VW conversions.
the cause was too heavy a prop bringing the harmonic vibration speeds down into the operating rpm range.
overheating heads were experienced by one of the european motorcycle manufacturers.
the solution was to move the transition to the exhaust pipe closer in to the valve seat. this took heat from the head and pushed it into the pipe which could be cooled more easily and wasn't as temperature critical.
the core challenge in fixing problems is to identify the actual cause.
sometimes it needs the calling in of some smart boffinry. (that's a hint jabiru.)
crankcase through bolts used to fail on VW conversions.
the cause was too heavy a prop bringing the harmonic vibration speeds down into the operating rpm range.
overheating heads were experienced by one of the european motorcycle manufacturers.
the solution was to move the transition to the exhaust pipe closer in to the valve seat. this took heat from the head and pushed it into the pipe which could be cooled more easily and wasn't as temperature critical.
the core challenge in fixing problems is to identify the actual cause.
sometimes it needs the calling in of some smart boffinry. (that's a hint jabiru.)
Join Date: Jul 2005
Location: Queensland
Posts: 686
Likes: 0
Received 0 Likes
on
0 Posts
so Jabiru decided to dumb the engine down by fitting hydraulic lifters. The mass of the solid lifter is tiny in comparison. Ergo large mass smashing back and fort
I don't think mass had much to do with hydraulic valve problems. I think it was more a cam shape problem. It allowed the valves to shut too quickly or open too slowly or something like that. Forgive me but it was a while ago that I heard the explanation.
I fully agree about the need to control cooling and the need for accurate gauges.
Thank goodness that I've never flown with Creampuff!
(Along with his often demonstrated limited knowledge of how his IO520 actually works.)
There should be a law against people like him!
(Along with his often demonstrated limited knowledge of how his IO520 actually works.)
There should be a law against people like him!
LOP/ROP arguments about Jabirus are missing the point; with a Bing Carburettor and no EGT probes, who would have the faintest idea of what is going on.
Furthermore, Jabirus issues have nothing to do with mixture and everything to do with basic mechanical design.
1. The lack of a head gasket makes the head/cylinder joint a weakness because of differential expansion and permanent distortion.
2. (1) means that valve clearance setting is problematic. That leaves valves vulnerable to sticking/burning.
3. Through-bolts are marginally too weak and clamping forces are not enough to prevent fretting without permanently deforming the heads (see 1.).
4. Perhaps most importantly, Jabirus use of fully CNC machined heads puts a limit on the number of cooling fins. the tool that cuts them has a maximum width/ depth ratio that cannot be exceeded ( its a machining thing) so there are not as many fins (more surface area) as ideally should be present to cool the head. A cast head with more and finer fins would be better from a cooling point of view.
Sadly, the last aerospace quality foundry that could have cast such heads was at CAC ands shut down due to stupid accounting practices that falsely indicated it was unprofitable.
5. Finally, the factory management either can't afford or won't contemplate product improvement, leaving them with no choice but to close business and sell it to someone who has the funds and attitude necessary for product improvement and professional customer service.
Jeez I'm glad I forgot about this motor.
Furthermore, Jabirus issues have nothing to do with mixture and everything to do with basic mechanical design.
1. The lack of a head gasket makes the head/cylinder joint a weakness because of differential expansion and permanent distortion.
2. (1) means that valve clearance setting is problematic. That leaves valves vulnerable to sticking/burning.
3. Through-bolts are marginally too weak and clamping forces are not enough to prevent fretting without permanently deforming the heads (see 1.).
4. Perhaps most importantly, Jabirus use of fully CNC machined heads puts a limit on the number of cooling fins. the tool that cuts them has a maximum width/ depth ratio that cannot be exceeded ( its a machining thing) so there are not as many fins (more surface area) as ideally should be present to cool the head. A cast head with more and finer fins would be better from a cooling point of view.
Sadly, the last aerospace quality foundry that could have cast such heads was at CAC ands shut down due to stupid accounting practices that falsely indicated it was unprofitable.
5. Finally, the factory management either can't afford or won't contemplate product improvement, leaving them with no choice but to close business and sell it to someone who has the funds and attitude necessary for product improvement and professional customer service.
Jeez I'm glad I forgot about this motor.
Sunfish I call you bluff on items 1 thru 4 B. You have been reading too many (other) uneducated posts on forums.
Eg,
You are saying that weak thru-bolts are causing fretting heads?
The jury is out on number 5.
Perhaps stick with whatever it is that you believe you do well in life versus Jabiru design engineering.
Eg,
You are saying that weak thru-bolts are causing fretting heads?
The jury is out on number 5.
Perhaps stick with whatever it is that you believe you do well in life versus Jabiru design engineering.
LOP/ROP arguments about Jabirus are missing the point; with a Bing Carburettor and no EGT probes, who would have the faintest idea of what is going on.
I am not going to expand on this because it is common to all aircraft engines and is the same old story, fit the engine analyzer now. In Jabiru's case they should just inject the engine, either electronically or manually.
That way you can easily do this:
Anyone who knows [Jabiru engines] will tell you to keep the CHT's below 110 C and the heads are a heap more stable.
Once you can control and measure EGT, you can set a mixture (and RPM) that doesn't result in the cylinders getting the hardest beating you can give them. Once you can control and measure a couple of key variables, you have a better chance of working out what's breaking, and why.
Considering your 'confident' attitude I assume you have the data for that. Care to share so we can critique?
Join Date: Jul 2005
Location: Queensland
Posts: 686
Likes: 0
Received 0 Likes
on
0 Posts
Considering your 'confident' attitude I assume you have the data for that. Care to share so we can critique?
Oh except millions of documented operational hours on turbo compound radials and other big bore radials as used before the jet age.
The turbo compound engines fitted to B29s and Super Constellations went from 200 hrs- 600 hrs time between overhauls when run rich of peak EGT to 3,600 hrs time between overhauls when run lean of peak.
But hey what's a few million hours of documented evidence compared to pet theories , established old wives tales and information supplied by your lame / instructor God. It's the vibes.
Of course we should also ignore data generated at GAMI which uses the most advanced aircraft piston engine diagnostic machine in the world.
Join Date: Nov 2012
Location: The Last Resort
Age: 52
Posts: 154
Likes: 0
Received 0 Likes
on
0 Posts
Please read again
I don't believe reducing unnecessary maintence and the potential for inducing oil leaks every 25 hours is 'dumbing down'. I for one encouraged Rod to fit hydraulic lifters every time I saw him. Even my old 55 Vauxhall had hydraulic lifters. People who like adjusting tappets need to get a life.
Read my post again, once temperatures are stable there is no need to adjust the valve clearance. I cant be bothered adjusting valves either. Through bolt tensions also stabilize. In the case of Jabiru engines I will take the solid lifter every time.
I'd like to point out, having done a bit of study on the subject in a past project, that fin surface area isn't the be-all and end-all of air cooling. Unfortunately, the problem is more complex than that because fin depth and width determines how much air passes through which is also a huge factor in heat transfer.
Bottom line is increasing the number of fins and decreasing fin spacing isn't necessarily a way to improve heat rejection and can reduce it instead.
Surprisingly, the best method to improve cooling appears to be to cowl each cylinder to precisely control the airflow through the fins, since air that passes over the cylinders rather than through the fins isn't doing much cooling at all. You have to do this in conjunction with eliminating cooling leakage though otherwise the general problem remains.
Oh, and crankcase problems? I heard from more than one person that the 912/914 series had crankcase fretting problem at some point during their history too.
Bottom line is increasing the number of fins and decreasing fin spacing isn't necessarily a way to improve heat rejection and can reduce it instead.
Surprisingly, the best method to improve cooling appears to be to cowl each cylinder to precisely control the airflow through the fins, since air that passes over the cylinders rather than through the fins isn't doing much cooling at all. You have to do this in conjunction with eliminating cooling leakage though otherwise the general problem remains.
Oh, and crankcase problems? I heard from more than one person that the 912/914 series had crankcase fretting problem at some point during their history too.
In the interests of disclosure, I run a 3300 Camit engine. They appear to have addressed all of the major Jabiru weaknesses and I’m very happy with the product. It has a starfish collar securing the flywheel, 6061 alloy heads, thicker through bolts than standard Jabiru bolts, piston squirters (aka Lycoming), angled rocker cams, oil filter locking mechanism, updated valves, alternator and more, so it’s clear they are working on the “fixes” and have what appears to be a great product.
I replaced the standard 3300 Jabiru engine in my aircraft with the CAMIT and continued with the standard factory cooling ducts and cowl etc and it simply ran too hot. The LAME had been treating the symptoms by replacing heat affected parts such as heads , valves and cylinders but not the cause of the over-heating. I won’t blame the LAME for a minute as these decisions are usually owner based, but in this case I have no evidence of either.
Basically, the installation was not suitable for either of these engines. I then embarked on a process of monitoring my engine (6 x EGT and 6 x CHT) and eliminating the issues that were causing the heat.
Airflow, airflow and airflow were the fixes, plus a small mixture alteration to bring the temps within tolerance.
Constrained by using a bing carby, I tried a multitude of mixture settings using a series of needle and main jets, some of which were ridiculously over-sized. Nothing worked in getting the CHT’s down until I increased the airflow.
I now enjoy running average CHT temps of around 125 degrees (~257f) with my max takeoff temps having never exceeded 150c (300f). Max OAT thus far 36 deg.
If anyone is interested in some pictures of the modifications required to achieve these figures, please contact me. As for those suggesting that you can’t perform these modifications due to certification constraints, needs to simply discuss with the factory.
I replaced the standard 3300 Jabiru engine in my aircraft with the CAMIT and continued with the standard factory cooling ducts and cowl etc and it simply ran too hot. The LAME had been treating the symptoms by replacing heat affected parts such as heads , valves and cylinders but not the cause of the over-heating. I won’t blame the LAME for a minute as these decisions are usually owner based, but in this case I have no evidence of either.
Basically, the installation was not suitable for either of these engines. I then embarked on a process of monitoring my engine (6 x EGT and 6 x CHT) and eliminating the issues that were causing the heat.
Airflow, airflow and airflow were the fixes, plus a small mixture alteration to bring the temps within tolerance.
Constrained by using a bing carby, I tried a multitude of mixture settings using a series of needle and main jets, some of which were ridiculously over-sized. Nothing worked in getting the CHT’s down until I increased the airflow.
I now enjoy running average CHT temps of around 125 degrees (~257f) with my max takeoff temps having never exceeded 150c (300f). Max OAT thus far 36 deg.
If anyone is interested in some pictures of the modifications required to achieve these figures, please contact me. As for those suggesting that you can’t perform these modifications due to certification constraints, needs to simply discuss with the factory.
Sqwawk, when the heads get too hot, they deform at the joint with the cylinder. That much is known. The joint leaks after deformation which is why Jabiru goes to positively insane lengths to instruct owners regarding cylinder head tightening. This should be a non issue if a modern head gasket was employed, even without going to the modern practice of designing the joint to employ plastic deformation of the head bolts during the tightening process.
As for through bolt failures, the "Cosi Fan Tutti" argument (every one does it) begs the question.
As for through bolt failures, the "Cosi Fan Tutti" argument (every one does it) begs the question.
This is very important and interesting stuff, S7700.
With your series of needle and main jets, what were the corresponding EGTs?
Were you able to determine where each set put you on the lean curve?
With your series of needle and main jets, what were the corresponding EGTs?
Were you able to determine where each set put you on the lean curve?
Too advanced for me there creamie.
I was simply adjusting mixture to achieve ideal egt figures based on >75% power and <75% power. Needle crafting was included to modify temps through the rpm range. It's not scientific and goes nowhere near LOP type discussions.
I proved (as the manufacturer said at the time) that adding fuel would be an expensive way to cool the engine and wouldn't work anyway. To richen the mixture from 20 to 28 lph has nil impact on CHT.
If we were talking figures closer to 18-19 lph then we would be talking more about LOP etc. I started well above this I suspect.
As for corresponding EGT's... The EGT's started high and gradually reduced as needle and main increased. Once you get too rich it's clear you are wasting fuel.
A $1000-$1500 investment in engine monitoring electronics is highly recommended.
Last check was 22lph. Without a TBI, LOP operations are just a dream.
Sunfish. Keeping head temps low is the solution, not fitting a gasket which will simply mask the problem.
I was simply adjusting mixture to achieve ideal egt figures based on >75% power and <75% power. Needle crafting was included to modify temps through the rpm range. It's not scientific and goes nowhere near LOP type discussions.
I proved (as the manufacturer said at the time) that adding fuel would be an expensive way to cool the engine and wouldn't work anyway. To richen the mixture from 20 to 28 lph has nil impact on CHT.
If we were talking figures closer to 18-19 lph then we would be talking more about LOP etc. I started well above this I suspect.
As for corresponding EGT's... The EGT's started high and gradually reduced as needle and main increased. Once you get too rich it's clear you are wasting fuel.
A $1000-$1500 investment in engine monitoring electronics is highly recommended.
Last check was 22lph. Without a TBI, LOP operations are just a dream.
Sunfish. Keeping head temps low is the solution, not fitting a gasket which will simply mask the problem.
Great stuff.
As a matter of interest, what was the impediment to trying sets that made the mixture leaner and leaner (noting that any EGT leaner than about 40 degrees rich of peak will start reducing CHT).
As a matter of interest, what was the impediment to trying sets that made the mixture leaner and leaner (noting that any EGT leaner than about 40 degrees rich of peak will start reducing CHT).
Impediments were, time, effort, cost, skill and a Bing. Without a mixture control there would be little to no point I believe.
Bugga!
I just thought that if there were sets of needles and main jets that made the mixture increasingly rich, there might also be sets that went the other way…
Who knows: You might have ended up with ‘rough’ running, due to differences in the mixtures being delivered to each cylinder through a messy induction system, but then again, you might not have.
Looks like we’ll just have to wait for Jabiru to do what Oracle1 is suggesting they should do (as well as ensure sufficient air flow for cooling in any event).
I just thought that if there were sets of needles and main jets that made the mixture increasingly rich, there might also be sets that went the other way…
Who knows: You might have ended up with ‘rough’ running, due to differences in the mixtures being delivered to each cylinder through a messy induction system, but then again, you might not have.
Looks like we’ll just have to wait for Jabiru to do what Oracle1 is suggesting they should do (as well as ensure sufficient air flow for cooling in any event).
It's all about choice, one can chose to risk beyond reasonable doubt ones life behind a Jabba donk or one can simply increase his/her chances of staying alive by choosing more known safe options:-)
Wmk2
Wmk2
Join Date: Apr 2008
Location: Australia
Posts: 490
Likes: 0
Received 0 Likes
on
0 Posts
If you want to lean your (Experimental) Jabiru you can:
HACman Mixture Control Order Page
I know nothing about the company, and I wouldn't fit it to my own aircraft. I reckon it would be a gamble whether leaning improved the current situation or made it worse. If I had to put money on it I would guess it would make it worse, but I don't mind being proved wrong.
If you try it, please report back...
HACman Mixture Control Order Page
I know nothing about the company, and I wouldn't fit it to my own aircraft. I reckon it would be a gamble whether leaning improved the current situation or made it worse. If I had to put money on it I would guess it would make it worse, but I don't mind being proved wrong.
If you try it, please report back...
But Andrew, how do you know where on the lean curve each of the cylinders on these engines is running now?
If leaning from X to Y on the lean curve could cause damage to an engine, what if the engine is already running at Y?
If leaning from X to Y on the lean curve could cause damage to an engine, what if the engine is already running at Y?